Water Wastage Detection in Smart Homes Through IoT and Machine Learning

Promoting sustainable water usage is a critical imperative across all sectors of society. Households are no exception since a significant portion of water is wasted daily due to inefficient appliances or improper habits. Thus, there is a need for innovative solutions that not only improve water utilization but also raise residents' awareness about this issue. This paper presents a promising solution leveraging the Internet of Things (IoT) and Machine Learning (ML) techniques to detect water wastage stemming from sink usage automatically. We have designed and developed a low-cost prototype equipped with an array of sensors, including a microphone, an ultrasonic sensor, and a PIR, to monitor sink usage. A deep learning model based on Gated Recurrent Units (GRU) has been trained to classify the wastage events. To validate our concept, we have gathered a small dataset relative to nine common daily water usage activities through the IoT prototype. Our preliminary findings demonstrate the feasibility of our solution, with an average accuracy exceeding 90% in detecting wastage events.